A Low Positive Rate Attendant for Sickle Cell Anemia Screening in Health and Medicine

African countries are facing a major public health problem due to hemogloblinopathies, which is the most common blood disorder in the world. In sub-Saharan Africa, 80% of the worldwide affected people (around 700 million) are living with sickle cell disease. Sickle cell anemia primarily affects children and teenagers and it unfortunately requires a life-long treatment from birth. Research reveals that in areas with no clinical follow-up, over 50% of children with sickle cell disease pass away before turning five years old. Health professionals assert that, depending on the patient's age and disease stage, early diagnosis and routine follow-up can prevent this scourge. However, standard and hand-operated diagnosis approaches, such as those used in Senegal and other sub-Saharan African nations, lack objective solutions and frequently require laborious decomposition operations that prone visual assessment errors. One of the most recent applications of AI is in medical decision-making, aiding doctors in diagnosis and treatment decisions. While interested in CNN learning model, ResNets, we propose an “Attendant of Screening for Sickle Cell Anemia” named (ASSCA) as a part of diagnosis decision-making. The ASSCA framework is able to accurately detect sickle cell disease and its development stage with a low false positive rate based on deep analysis of human blood smears images. In addition, based on knowledge of medical concerns, ASSCA gives better ways to measure disease assessment for rapid diagnosis decisions related to appropriate care. A key processing step is introduced to tackle classification challenges due to complex cell shapes, overlapped cells, and image color range. With a suitable optimizer (SGD-M), ASSCA gives an outstanding mAP measure of 98.81 % with a low false positive rate.